EFFECT OF DRYING ON COLOR, PROXIMATE COMPOSITION
AND DRYING KINETICS OF SLICED CHESTNUTS
TERESA DELGADO1,2, JOSÉ ALBERTO PEREIRA1, SUSANA CASAL2,3and ELSA RAMALHOSA1
1Mountain Research Centre (CIMO) – School of Agriculture, Polytechnic Institute of Bragança, Campus de Sta Apolónia, Apartado 1172, Bragança
5301-855, Portugal
2LAQV-REQUIMTE, Faculty of Pharmacy of University of Porto, Rua de Jorge Viterbo Ferreira, n.° 228, Porto 4050-313, Portugal
3Corresponding author.
TEL:+351-220428638;
FAX:+351-226093390;
EMAIL: sucasal@ff.up.pt
Received for Publication January 6, 2015 Accepted for Publication May 14, 2015 doi:10.1111/jfpe.12244
ABSTRACT
In the present work, dried sliced chestnuts (Judia and Longal varieties), product with an increased shelf life, low calorie and gluten-free contents, were prepared. The effect of air convective drying on the drying kinetics, color and proximate composition of sliced chestnuts was evaluated. Even though significant differences in nutritional composition were found between both varieties at the beginning, the drying behaviors were similar; however, Judia dried at a slightly faster rate than Longal. The use of Page, two-term, and modified Henderson and Pabis models fitted well the experimental data (adjusted R2> 0.999). With drying, slight variations in color were observed for both varieties and only moisture content decreased significantly.
The obtained product retained all chemical composition, and due to the low caloric value (367 kcal/100 g product), low fat and gluten-free contents of chest-nut slices, this can be an interesting substitute to other high-calorie snacks avail-able in the market.
PRACTICAL APPLICATIONS
The chestnut fruit is increasingly popular among consumers. The fruit is usually sold fresh or frozen while smaller fruits are generally rejected by industries. So, it is very important to find alternatives to valorize these fruits. Moreover, consumers search for healthy and easy-to-consume food. Chestnut follow these requisites, being a nut with interesting properties due to its low fat content, high levels of starch (sugar of slow absorption) and significant amounts of fibers. Furthermore, it is a gluten-free nut, ideal for celiac patients. On the other hand, the majority of snacks in the market are rich in fat and are made from wheat flour. So, the devel-opment of snacks based on chestnut would be innovative. This study intends to provide information on the effect of drying on color, nutritional composition and drying kinetics of sliced chestnut in order to obtain a healthy and low-calorie content snack.
INTRODUCTION
Consumers’ search for healthy and easy-to-consume food is increasing. Chestnut follows these requisites, being a nut with interesting properties due to its low fat content, high levels of starch (sugar of slow absorption) and significant amounts of fibers (Borges et al. 2008). Furthermore, it is a gluten-free nut, ideal for celiac patients. On the other hand, the majority of snacks in the market are rich in fat
and are made from wheat flour (that contains gluten). So, the development of snacks based on chestnut would be innovative and will allow the development of new prod-ucts to persons with sensitivity to gluten. The production of dried slices of chestnuts emerged as a possible and healthy option, with low-calorie content.
Several studies on hot air drying have been performed in chestnuts, namely on drying kinetics (Moreira et al. 2005; Guiné and Fernandes 2006; Cletus and Carson 2008),
drying characteristics and energy requirement for dehydra-tion (Koyuncu et al. 2004), the effect of drying temperatures on morphological, chemical, thermal and rheological prop-erties in chestnut flours (Correia et al. 2009; Correia and Beirão-da-Costa 2012; Moreira et al. 2013), the effect of drying followed by rehydration on different properties of chestnuts (Attanasio et al. 2004; Moreira et al. 2008, 2011) and how starch is affected by different drying methods (Zhang et al. 2011).
Until now, most of these studies have been carried out with peeled and unpeeled whole fruits. Only one study using prismatic chestnut samples (10× 10 × 15 mm), sub-jected to convective air drying at 65± 0.5C has been per-formed (Moreira et al. 2011). Still, only the chestnut cellular tissue was analyzed with significant changes in cell size as drying process progressed. So, until now few results exist on the role of drying in sliced chestnuts properties. Thus, the aim of our work was to assess the effect of hot-air convec-tive drying in physical and proximate composition of chest-nuts cut in slices along drying time in order to analyze the effect of thermal processing on them.
MATERIALS AND METHODS
Plant Material
Two chestnut varieties were used in this study, namely Longal and Judia, the most common in the region. The nuts were acquired directly to chestnut producers of Macedo de Cavaleiros (Longal variety) and Vinhais (Judia variety) at north-east of Portugal (50 kg each variety) in November
2012 and stored in cold chambers (4± 1C) (maximum 1 month) until the analyses were carried out.
Drying Experiments
After removing carefully the exterior shell with a knife, chestnuts were sliced with approximately 4–6 mm of thick-ness (Fig. 1 at 0 h). Then, for fixed time periods (1, 2, 4, 6, 8 and 10 h), around 150 g of chestnut slices were dried in a tray dryer (Armfield, Ringwood, England) at 50C (Fig. 2). This temperature was chosen because it was the maximum allowable by the equipment and it was a common tempera-ture found on other drying studies of fruits (Koyuncu et al. 2004; Correia et al. 2009; Correia and Beirão-da-Costa 2012). The tray dryer consisted of one fan and an electric heating element (maximum power 3 kW). The hot air passed through the central section of the duct, where the tray with the material to drying is fixed, going out to the atmosphere. The tray was connected to a decimal scale and the data acquisition was recorded in a computer through the Windows Hyperterminal software (Hilgraeve, Monroe, Michigan, USA).
The air velocity was measured with a portable anemom-eter (Airflow, LCA 6000, Buckinghamshire, England) and kept constant at 1.2± 0.1 m/s throughout experiments. The control (fresh chestnuts) and drying samples (1, 2, 4, 6, 8 and 10 h) were frozen and freeze-dried in order to determine their moisture contents and their proximate composition were expressed in dry basis. All drying experiments at each time period were performed in triplicate. h 0 1 = t h 0 = t s e i t e i r a V Longal Judia
FIG. 1. CHESTNUTS OF LONGAL AND JUDIA
VARIETIES AT 0 AND 10 H AFTER DRYING AT 50C
The drying process was followed by weighting the chest-nut slices at regular time intervals (40 s) with an accuracy of± 0.1 g, being the moisture ratios (MR) determined by Eq. (1). MR W W W W e e = − − 0 (1)
where W, W0and Werepresent the instantaneous, initial and
equilibrium dry basis water contents (kg water/kg dry basis), respectively. We was determined as the asymptotic
value of the function fit of the experimental points at the final stage of drying.
Several mathematical models were tested to fit the mois-ture ratio versus time; however, only the three best models are presented, whose equations are shown in Table 1.
The model parameters were determined by SPSS software (Version No. 20.0, IBM Corporation, Armonk, New York, USA) and the suitability of the fits was evaluated by the fol-lowing parameters: SSE y i y el i i n =
(
−)
=∑
exp, mod , 2 1 (2) SST y yaverage i n =(
−)
=∑
exp 2 1 (3) R SSE SST 2= −1 (4) 1 2 3 4 6 5 8 7FIG. 2. PHOTOGRAPH AND SCHEMATIC FIGURE OF THE TRAY DRYER
(ARMFIELD): FAN (1); HEATER (2); CONTROL PANEL (3); DECIMAL BALANCE (4); TEMPERATURE AND HUMIDITY SENSORS (5) AND (6); TRAY FOR SAMPLES (7); COMPUTER (8)
TABLE 1. DRYING CURVE MODELS AT 50C FOR TWO CHESTNUT VARIETIES
Variety
Page model Two-term model Modified Henderson and Pabis model
MR= exp(−k × tn) MR= a × exp(−k × t) + b × exp(−g × t) MR= a × exp(−k × t) + b × exp(−g × t) + c × exp(−h × t)
Judia k= 0.3727 a= 0.115 a= 0.8977 n= 0.8973 k= 2.566 b= 0.5535 SSE: 0.02617 b= 0.8972 c= −0.4393 R2: 0.9995 g= 0.2925 g= 2.286 Adjusted R2: 0.9995 SSE: 0.00343 h= 2.217 RMSE: 0.00540 R2: 0.9999 k= 0.2926 Adjusted R2: 0.9999 SSE: 0.00346 RMSE: 0.001956 R2: 0.9999 Adjusted R2: 0.9999 RMSE: 0.00197 Longal k= 0.3462 a= 0.8619 a= 0.2803 n= 0.8452 k= 0.237 b= 0.5799 SSE:0.03826 b= 0.1446 c= 0.1469 R2: 0.9992 g= 2.218 g= 0.2363 Adjusted R2: 0.9992 SSE: 0.00164 h= 2.189 RMSE: 0.00652 R2: 1.000 k= 0.2363 Adjusted R2: 1.000 SSE: 0.00104 RMSE: 0.00135 R2: 1.000 Adjusted R2: 1.000 RMSE: 0.00108 RMSE, calculated by Equation (6); SSE, calculated by Equation (2).
R adj SSE n SST where n m 2 1 1 = − × −
(
)
×υ υ= − (5) RMSE= SSE υ (6)where yexp,i and ymodel,i are the experimental and predicted
values for the i observation, respectively, n is the number of observations and m is the number of parameters in each model (Yaldýz and Ertekýn 2001; Togrul and Pehlivan 2003).
Physicochemical Characterization
Color. The color of chestnut slices was determined during
the drying experiments by a Minolta CR-400 colorimeter in CIELab color space, through the coordinates: L*, a* and b*, using the Spectra Magic Nx software (version CM-S100W 2.03.0006, Konica Minolta Company, Osaka, Japan) described in a previous study (Delgado et al. 2014). The C* (chroma or saturation) and h* (hue angle) were also calcu-lated by the following equations:
C*= a*2+b*2 (7) h b a * * * =arctan⎛⎝ ⎞⎠ (8)
chroma (C*) measures the purity or saturation of the color, while hue (h*) denotes the subtle distinction or variation in color (Rajasekar et al. 2012). In order to analyze the changes on color along the drying process, this parameter was deter-mined at the beginning (color of fresh chestnut, considered as reference) and after the drying process on 60 slices. So, theΔL*, Δa*, Δb*, ΔC* and Δh* were determined by the dif-ference of the values at the end and the beginning of the drying process. Moreover, the total color difference (ΔE*) was also calculated according to
ΔE*=
(
ΔL*)
2+(
Δa*)
2+(
Δb*)
2 (9)Proximate Composition
The samples were analyzed for proximate composition (moisture, proteins, fat and ash) using Association of Offi-cial Analytical Chemists (AOAC) procedures (AOAC 1995). All reagents were of analytical grade and purchased from Sigma-Aldrich Chemical Co. (St Louis, MO). Crude protein content of the samples was estimated by the macroKjeldahl method (VELP SCIENTIFICA, Usmate Velate, Italy), using a conversion factor of 5.3 (Borges et al. 2008; Mendes de Vasconcelos et al. 2009); crude fat was determined by extracting 5 g of sample with petroleum ether for 24 h, using a Soxhlet apparatus (P Selecta, Abrera, Barcelona,
Spain) and ash content was determined by incineration at 550C (Lenton Thermal Designs Ltd, Hope Valley, UK) for 4 h. Neutral detergent fiber (NDF) and acid detergent fiber (ADF) were determined by the method described by Goering and Van Soest (1970). Total carbohydrates were calculated by difference as described by FAO (2003). Total energy was calculated according to Atwater system. All determinations were made in duplicate, comprising six values for each drying treatment performed in triplicate.
Statistical Methods
The statistical analysis was performed on SPSS software (Version No. 20.0). Comparisons were carried out at 95% confidence by application of analysis of variance and when significant differences between samples were observed, the Tukey’s Honestly Significant Difference test (Tukey’s HSD post hoc) was applied.
RESULTS AND DISCUSSION
Drying Kinetics
The drying curves of the two chestnut varieties, previously cut in slices, are shown in Fig. 3. These curves are typical drying curves with two phases. At the beginning the drying rate decreased rapidly, followed by a slowly decrease. Both varieties showed similar behaviors along the drying time. Small differences were detected on the initial phase, prob-ably due to their different initial moisture contents, having Judia higher moisture content (52.7%) than Longal variety (48.5%) (Fig. 3A). However, by analyzing Fig. 3B we can see that to reach the same moisture ratio (e.g., 0.20), the Judia variety lost water more easily than Longal, since a shorter period of time was needed (around 5 h) for the former, while Longal needed a little longer than 6 h to achieve the same value.
Three different mathematical models, namely Page (with one exponential), two-term (with two exponentials) and modified Henderson and Pabis (with three exponentials) models were tested to describe the behavior of chestnuts along drying. The model parameters and the statistics used to evaluate the models suitability are presented in Table 1. It was found that all models described well the experimental data. From the statistical parameters calculated, it was observed that the quality of the fit was good for both chest-nut varieties subjected to drying (Table 1), with R2 and adjusted R2in the range of 0.9992 to 1. As all values were near 1, these models described very well the experimental data. SSE and RMSE also presented low values (close to zero, as desired), varying between 0.00104 to 0.0383 and 0.00108 to 0.00652, respectively. In general terms, after observing the statistical parameters of the three models, a
slight improvement was obtained when using two exponen-tial terms compared with Page model with a single expo-nential. However, with this simple model, good results were obtained. So, this model may be used in the future to predict very satisfactorily the drying curves of chestnut slices.
Effect of Hot-Air Convective Drying on Sliced Chestnut Color
Color is a highly appreciated property by chestnut consum-ers. Thus, it is important to analyze the effect of drying in
this physical characteristic. The results concerning the color changes throughout drying time are detailed in Table 2. Sig-nificant interactions between variety and drying time were always observed (P< 0.001).
An increase inΔL* absolute values was observed along drying until 4 h for Longal and 6 h for Judia varieties. After this period, the values remained almost constant. Moreira
et al. (2005) observed a decrease on this parameter when
analyzing the L* values of peeled and cut chestnuts after drying at 45, 55 and 65C in line with that observed in the present work. In our study, a decrease in the L* values was
A
B
tJudia tLongal
FIG. 3. DRYING RATES IN TERMS OF MOISTURE CONTENT (A) AND MOISTURE RATIO (B) ALONG DRYING TIME FOR LONGAL AND JUDIA
VARIETIES
TABLE 2. COLOR PARAMETERS ALONG AIR CONVECTION DRYING AT 50C FOR TWO CHESTNUT VARIETIES, LONGAL AND JUDIA
Parameter
Air convection drying P
t= 1 h t= 2 h t= 4 h t= 6 h t= 8 h t= 10 h Variety× Time
ΔL*
Longal −1.64 ± 3.34c,A −3.11 ± 4.69b,A −5.22 ± 5.70a,A −4.70 ± 4.84a,A −5.21 ± 3.97a,A −4.16 ± 4.27a,b,A <0.001
Judia −1.59 ± 3.29c,A −2.44 ± 3.21c,A −3.76 ± 3.92b,B −5.32 ± 4.04a,A −4.84 ± 5.01a,b,A −5.81 ± 4.13a,B
Δa*
Longal 0.56± 0.74a,A 0.72± 0.91a,b,A 0.92± 1.53b,c,A 1.39± 1.00d,A 1.13± 1.09c,d,A 1.19± 0.85c,d,A <0.001
Judia 0.53± 0.90a,A 0.64± 0.92a,b,A 0.88± 0.97b,c,A 0.86± 1.08b,c,B 1.16± 0.95c,d,A 1.34± 0.78d,A
Δb*
Longal −0.54 ± 4.28a,b,A 0.46± 4.33b,A 0.30± 5.65b,A −1.54 ± 4.62a,A −0.08 ± 4.16a,b,A −1.44 ± 4.55a,A <0.001
Judia 0.39± 4.38b,c,A −0.04 ± 5.01b,c,A −0.64 ± 4.54a,b,c,A 0.75± 5.29c,B −1.00 ± 5.13a,b,A −1.56 ± 4.17a,A
ΔC*
Longal −0.59 ± 4.30a,b,c,A 0.40± 4.34c,A 0.25± 5.65c,A −1.66 ± 4.62a,A −0.17 ± 4.19b,c,A −1.54 ± 4.58a,b,A <0.001
Judia 0.34± 4.41b,c,A −0.10 ± 5.03b,c,A −0.72 ± 4.57a,b,c,A 0.68± 5.31c,B −1.09 ± 5.14a,b,A −1.66 ± 4.18a,A
Δh*
Longal −1.31 ± 1.56c,A −1.88 ± 2.08b,c,A −2.39 ± 4.05a,b,A −3.20 ± 2.36a,A −2.91 ± 2.49a,A −2.67 ± 1.83a,b,A <0.001
Judia −1.33 ± 1.69d,A −1.52 ± 1.87c,d,A −2.05 ± 1.84b,c,A −1.98 ± 2.02b,c,B −2.47 ± 2.06a,b,A −3.01 ± 1.91a,A
ΔE*
Longal 5.02± 2.82a,A 5.93± 4.09a,b,A 8.10± 5.41d,A 7.58± 3.82c,d,A 6.80± 4.05b,c,A 6.99± 3.37b,c,d,A <0.001
Judia 4.83± 3.20a,A 5.69± 3.18a,b,A 6.42± 3.29b,B 7.84± 3.72c,A 7.97± 3.78c,B 7.83± 3.41c,B
Mean± standard devaition with different small letter (a–d) superscripts on the same row are significantly different (P < 0.05). Mean ± standard
also detected, explaining the negative values obtained for the ΔL* (Table 2). Moreover, in the present work, differ-ences between varieties were observed at the beginning (0 h) (87.3 for Longal and 86.5 for Judia varieties) and after 6 or more hours of drying. Judia always showed lower L* values than Longal (data not shown), indicative of lower lightness.
There was also a significant increase inΔa* values (which indicates higher red proportions) along drying for both varieties. For Longal variety, theΔa* values remained con-stant after 6 h, while for Judia theΔa* values kept on con-stant after 8 h. Moreira et al. (2005) also observed an increase in a* values of both peeled and cut chestnuts after drying at 45, 55 and 65C.
Regarding Δb* and ΔC* parameters, slight variations were observed along drying for both varieties. For Longal variety, after 10 h of drying, the yellowness variation (Δb* value) and chroma variation (ΔC*) were not significantly different to 1 h of drying. On contrary, a significant differ-ence was observed for Judia.
Regarding the hue (h*) values, both varieties presented a yellow predominant color (h* values around 90°) in line with the higher effect of b* component than a*, due to the higher values of the former (data not shown). However, along drying time, the variation increased in absolute terms for both varieties.
Total color differences (ΔE*) were evaluated along drying and values higher than zero were always observed, suggest-ing that chestnuts color changed along dehydration; however, this variation did not result in a different color becauseΔE* values were less than 12 (Cecchini et al. 2011) (Fig. 1). For Longal variety, the highest variation on ΔE* values was observed at 4 h of drying, remaining almost con-stant after that time. For Judia, a more concon-stant increase in this parameter was observed, remaining stable after 6 h. These results may be related to the occurrence of enzymatic browning reactions due to polyphenol oxidase activity during drying, as well as to nonenzymatic browning reac-tions, namely Maillard and caramelization reactions. In fact, Judia variety presents slight higher values of sugars and pro-teins than Longal that may favor the occurrence of such reactions, as explained subsequently, and supported by the increased redness (a*), typical color of these browning products.
Effect of the Hot-Air Convective Drying on Sliced Chestnut Proximate Composition
The results obtained for proximate composition of the two chestnut cultivars along the drying time are shown in Table 3. In almost all situations, significant interactions between variety and drying time were observed (P< 0.05),
except for crude protein and NDF. However, for these parameters the individual effects of chestnut variety and/or drying time were significant.
Water is the predominant component in fresh chestnuts (0 h), responsible for its high perishability. Significant dif-ferences on moisture contents were found between both varieties, with a mean of 48.5± 0.4 g/100 g fresh weight for Longal and 52.7± 1.4 g/100 g fresh weigh for Judia. These results were in accordance with the Spanish and Portuguese cultivars analyzed by Míguelez et al. (2004) (48.37 to 59.35 g/100 g fresh weight), Pereira-Lorenzo et al. (2006) (40.3 to 60.1 g/100 g fresh weight), Borges et al. (2008) (46.3 to 53.3 g/100 g fresh weight) and Correia et al. (2009) (47.9 to 48.2 g/100 g fresh weight). As expected, the moisture con-tents decreased along drying and stabilized without signifi-cant variations after 8 h for Longal and 6 h for Judia. When comparing both varieties, the significant differences observed at the beginning (0 h) disappeared, supported by a higher moisture loss in Judia, confirming the differences on drying kinetics referred previously.
Significant differences were always found in ash content between the two varieties, with an average of 1.76± 0.02 g ash/100 g dry weight for Longal and 2.25± 0.12 g ash/100 g dry weight for Judia at the beginning of the drying experi-ments (0 h). The differences between varieties may be related to genetic differences, altitude and soil type, among others, as proposed by Pereira-Lorenzo et al. (2006). In fact both varieties were from different locations; however, our values are in agreement with Ertürk et al. (2006) (1.02 to 3.22 g ash/100 g dry weight), Pereira-Lorenzo et al. (2006) (1.8 to 3.2 g ash/100 g dry weight), Borges et al. (2008) (1.60 to 2.20 g ash/100 g dry weight) and Correia et al. (2009) (1.9 to 2.1 g ash/100 g dry weight). Small variations were observed in Longal along drying, while no significant varia-tions were observed in Judia variety. This was expected as this property is not predictable to vary along dehydration.
The fat content in chestnuts is low, but still significantly higher in Longal (3.26± 0.11 g fat/100 g dry weight) than Judia (2.77± 0.45 g fat/100 g dry weight) before drying. These values are similar to previous studies such as Ertürk et al. (2006) (0.49 to 2.01 g fat/100 g dry weight), Pereira-Lorenzo et al. (2006) (1.7 to 4.0 g fat/100 g dry weight), Borges et al. (2008) (1.73 to 3.10 g fat/100 g dry weight) and Correia et al. (2009) (2.6 to 3.0 g fat/100 g dry weight) for Turkish, Spanish and Portuguese varieties. Along drying, the two varieties presented a similar fat content. Furthermore, the fat values determined at the beginning and after 10 h of the drying process were not sig-nificantly different, indicating that this parameter was almost unaffected by the thermal process.
Crude protein did not show significant differences along drying time but differed between the two chestnut varieties. At the beginning, Longal and Judia varieties presented crude
protein contents of 3.97± 0.20 and 5.43 ± 0.46 g protein/ 100 g dry weight, respectively. The higher protein content in Judia than Longal might support the higher formation of Maillard products during chestnut drying, as previously suggested. Our values are in agreement with those pub-lished by Míguelez et al. (2004) (6.02 to 8.58 g protein/100 g dry weight), Ertürk et al. (2006) (4.88 to 10.87 g protein/ 100 g dry weight), Pereira-Lorenzo et al. (2006) (4.5 to 9.6 g protein/100 g dry weight), Borges et al. (2008) (4.87 to 7.37 g protein/100 g dry matter) and Correia et al. (2009) (4.3 to 5.0 g protein/100 g dry weight). Míguelez et al. (2004) refer that differences between varieties may be related to differences in soil type, as soils with a greater amount of schist present higher protein content than granite-based soils.
ADF and NDF showed small variations along drying; however, generally, no significant differences were observed between 0 and 10 h. Judia showed a lower initial content in NDF than Longal variety. Nevertheless, no significant differ-ences between varieties were observed along drying. Also for ADF, similar results were obtained for both varieties for almost all drying times. Our ADF (3.45 and 3.72 g ADF/ 100 g dry weight) and NDF (5.93 and 9.74 g NDF/100 g
dry weight) values were similar to those obtained by Pereira-Lorenzo et al. (2006) (2.3 to 4.5 g ADF/100 g dry weight and 9.4 to 28.5 g NDF/100 g dry weight) and Borges
et al. (2008) (1.89 to 3.15 g ADF/100 g dry weight and 13.8
to 24.4 g NDF/100 g dry weight).
The level of carbohydrates in chestnuts was high (91.0± 0.2 g carbohydrates/100 g dry weight for Longal and 89.6± 0.4 g carbohydrates/100 g dry weight for Judia variety). Our results are a little higher than those published by Ertürk et al. (2006), 75.32 to 86.31 g carbohydrates/100 g dry weight, for several Turkish chestnut cultivars. Neverthe-less, after expressing the results in fresh weight, our results (46.9 g carbohydrates/100 g fresh weight for Longal and 42.4 g carbohydrates/100 g fresh weight for Judia variety) were similar to Barreira et al. (2009) for both varieties (44.1 g carbohydrates/100 g fresh weight for Longal and 42.1 g carbohydrates/100 g fresh weight for Judia variety). Concerning drying, no significant differences were observed between 0 and 10 h for each variety. So, carbohydrates deg-radation was not significant during the drying process, the remaining sugar contents were almost unchanged. However, significant differences between varieties were always found, having Longal variety slight higher values than Judia. TABLE 3. MOISTURE CONTENT AND NUTRITIONAL COMPOSITION ALONG AIR CONVECTION DRYING FOR TWO CHESTNUT VARIETIES, LONGAL
AND JUDIA
Parameter
Air convection drying P
t= 0 h t= 1 h t= 2 h t= 4 h t= 6 h t= 8 h t= 10 h Variety× Time
Moisture (g water/100g chestnut after drying)
Longal 48.5± 0.4f,A 36.7± 3.1e,A 30.8± 2.9d,A 17.3± 1.5c,A 16.6± 1.9b,c,A 11.8± 0.7a,b,A 11.0± 0.5a,A 0.013
Judia 52.7± 1.4e,B 38.9± 3.6d,A 32.0± 1.5c,A 17.6± 3.0b,A 11.5± 0.8a,B 11.4± 1.3a,A 10.3± 0.2a,A
Total ash (g ash/100g dry weight)
Longal 1.76± 0.02b,c,A 1.78± 0.03c,A 1.71± 0.03a,b,A 1.68± 0.03a,A 1.78± 0.05c,A 1.78± 0.02c,A 1.76± 0.04b,c,A 0.012
Judia 2.25± 0.12a,B 2.25± 0.05a,B 2.26± 0.08a,B 2.35± 0.15a,B 2.35± 0.07a,B 2.30± 0.06a,B 2.24± 0.07a,B
Crude fat (g fat/100g dry weight)
Longal 3.26± 0.11c,A 2.78± 0.40a,A 2.81± 0.12a,A 3.17± 0.08b,c,A 2.66± 0.15a,A 2.83± 0.16a,b,A 2.98± 0.15a,b,c,A 0.012
Judia 2.77± 0.45a,B 2.69± 0.41a,A 2.77± 0.24a,A 2.48± 0.32a,B 2.50± 0.11a,A 2.71± 0.33a,A 3.00± 0.19a,A
Crude protein (g protein/100g dry weight)
Longal 3.97± 0.20a,A 4.81± 0.15a,A 4.20± 0.32a,A 4.37± 0.12a,A 4.42± 0.38a,A 4.34± 0.52a,A 4.68± 0.28a,A 0.156
Judia 5.43± 0.46a,B 5.41± 0.42a,A 5.44± 0.30a,B 5.29± 0.65a,B 5.87± 0.30a,B 5.59± 0.53a,B 5.09± 0.03a,A
NDF (g NDF/100g dry weight)
Longal 9.74± 1.57a,A 16.78± 2.50b,c,A 18.08± 3.24c,A 17.87± 1.66b,c,A 16.73± 2.84b,c,A 17.78± 1.99b,c,A 13.49± 0.63a,b,A 0.057
Judia 5.93± 0.81a,B 13.52± 3.63b,A 17.35± 2.06b,c,A 16.49± 2.51b,c,A 15.46± 1.39b,c,A 18.00± 1.59c,A 15.76± 2.35b,c,A
ADF (g ADF/100g dry weight)
Longal 3.72± 0.43b,A 3.46± 0.37a,b,A 3.08± 0.37a,b,A 3.26± 0.40a,b,A 2.93± 0.32a,A 3.19± 0.42a,b,A 3.22± 0.34a,b,A 0.016
Judia 3.45± 0.30a,b,A 3.23± 0.22a,A 3.66± 0.46a,b,B 3.38± 0.33a,b,A 3.22± 0.44a,A 4.05± 0.62b,B 3.58± 0.52a,b,A
Total carbohydrates (g carbohydrates/100g dry weight)
Longal 91.0± 0.2a,b,c,A 90.7± 0.4a,b,A 91.3± 0.3c,A 90.8± 0.1a,b,c,A 91.1± 0.3b,c,A 90.8± 0.3a,b,A 90.6± 0.3a,A <0.001
Judia 89.6± 0.4a,B 89.6± 0.4a,b,B 89.5± 0.2a,B 90.2± 0.4b,B 89.3± 0.3a,B 89.4± 0.2a,B 89.7± 0.2a,b,B
Energetic value (kcal/100g dry weight)
Longal 412± 0.5c,A 410± 2a,b,A 410± 0.6a,b,A 412± 0.4c,A 409± 0.6a,A 410± 0.5a,b,A 411± 1b,c,A 0.003
Judia 409± 2a,B 408± 2a,A 409± 1a,B 407± 2a,B 408± 1a,B 408± 1a,B 409± 1a,B
Mean± SD with different small letter (a–f) superscripts on the same row are significantly different (P < 0.05). Mean ± SD with different capital letter (A–B) superscripts on the same column are significantly different (P< 0.05).
Concerning energetic value, Judia always presented lower values than Longal but the differences were almost negli-gible. At the beginning (0 h), Longal had 212 kcal/100 g while Judia had 193 kcal/100 g, similar to those obtained by Barreira et al. (2009) for the same varieties (193 kcal/100 g for Longal and 187 kcal/100 g fresh weight for Judia). Regarding drying, no significant differences were detected in Judia along time. Even though small variations were observed for Longal along drying, the energetic values on a dry basis at 10 h were identical to that presented at 0 h, indi-cating only slight changes in the macronutrients along the drying process.
When comparing the energetic value of dried chestnut slices per 100 g of product (367 kcal/100 g for both variet-ies) with other kind of snacks, such as fried peanuts and fla-vored chips, the former had approximately 32 to 40% less calories than the other snacks, suggesting it to be a healthy and less caloric substitute. On the other hand, the dried chestnut slices had a similar energetic value to dried apple (346 kcal/100 g). This product was developed by a Portu-guese enterprise and it had earned a nutrition award at 2013, being labeled as a healthy snack. The demand of this kind of product has nowadays been increasing due to greater consumer concern with health. Furthermore, we performed a preliminary sensory study where 10-member semi-trained panelists were asked to rate the overall accep-tance of the dried sliced chestnuts in a 10-cm nonstructured scale (0 means unacceptable and 10 means very acceptable). Good results were obtained with an average± standard deviation equal to 7.36± 1.41.
CONCLUSIONS
Even though slight differences in moisture content were found at the beginning between Longal and Judia varieties, among other properties, the drying behaviors were similar and the experimental data were well adjusted by Page, two-term and modified Henderson and Pabis models. This is important from the industrial point of view as not adjust-ments in the drying process are expected to be necessary for these two varieties, being their drying behaviors also easily predicted. Although color differences were detected after drying at 50C for 10 h, changes on chestnut macronutrients were generally not observed. The variability between varieties was higher than that observed along drying for each variety. Due to their low caloric and gluten-free contents, dried chestnut slices appear as an excellent substitute to other types of snacks.
ACKNOWLEDGMENTS
Teresa Delgado acknowledges the Fundação para a Ciência e Tecnologia (FCT) for the financial support through the
PhD grant – SFRH/BD/82285/2011, CIMO through the Project PEst-OE/AGR/UI0690/2014, REQUIMTE through the Project PEst-C/EQB/LA0006/2013 and POCTEP – Programa de cooperação Transfronteiriça Espanha – Portu-gal through the Project RED/AGROTEC – Experimentation network and transfer for development of agricultural and agro industrial sectors between Spain and Portugal.
REFERENCES
AOAC. 1995. Official Methods of Analysis, 16th Ed., Association of Official Analytical Chemists, Arlington, VA.
ATTANASIO, G., CINQUANTA, L., ALBANESE, D. and DI MATTEO, M. 2004. Effects of drying temperatures on physico-chemical properties of dried and rehydrated chestnuts (Castanea sativa). Food Chem. 88, 583–590. BARREIRA, J.C.M., CASAL, S., FERREIRA, I.C.F.R., OLIVEIRA,
M.B.P.P. and PEREIRA, J.A. 2009. Nutritional, fatty acid and triacylglycerol profiles of Castanea sativa Mill. cultivars: A compositional and chemometric approach. J. Agric. Food Chem. 57, 2836–2842.
BORGES, O., GONÇALVES, B., SOEIRO DE CARVALHO, J.L., CORREIA, P. and SILVA, A.P. 2008. Nutritional quality of chestnut (Castanea sativa Mill.) cultivars from Portugal. Food Chem. 106, 976–984.
CECCHINI, M., CONTINI, M., MASSANTINI, R., MONARCA, D. and MOSCETTI, R. 2011. Effects of controlled
atmospheres and low temperature on storability of chestnuts manually and mechanically harvested. Postharvest Biol. Technol. 61, 131–136.
CLETUS, A.B. and CARSON, J.K. 2008. Drying curves and apparent diffusivity of New Zealand chestnut variety “1015”. J. Food Eng. 85, 381–386.
CORREIA, P. and BEIRÃO-DA-COSTA, M.L. 2012. Effect of drying temperatures on starch-related functional and thermal properties of chestnut flours. Food Bioprod. Process. 90, 284–294.
CORREIA, P., LEITÃO, A. and BEIRÃO-DA-COSTA, M.L. 2009. The effect of drying temperatures on morphological and chemical properties of dried chestnuts flours. J. Food Eng. 90, 325–332.
DELGADO, T., PEREIRA, J.A., BAPTISTA, P., CASAL, S. and RAMALHOSA, E. 2014. Shell’s influence on drying kinetics, color and volumetric shrinkage of Castanea sativa Mill. fruits. Food Res. Int. 55, 426–435.
ERTÜRK, Ü., MERT, C. and SOYLU, A. 2006. Chemical composition of fruits of some important chestnut cultivars. Braz. Arch. Biol. Technol. 49, 183–188.
FAO. 2003. Food energy – Methods of analysis and conversion factors. in: FAO (Ed.), Food and Nutrition Paper 77. Rome, pp. 12–14.
GOERING, H.K. and VAN SOEST, P.J. 1970. Forage Fiber
Analyses – Apparatus, Reagents Procedures, and Some Applications, Agricultural Research Service – United States
GUINÉ, R.P.F. and FERNANDES, R.M.C. 2006. Analysis of the drying kinetics of chestnuts. J. Food Eng. 76, 460–467. KOYUNCU, T., SERDAR, U. and TOSUN, I. 2004. Drying
characteristics and energy requirement for dehydration of chestnuts (Castanea sativa Mill.). J. Food Eng. 62, 165–168. MENDES DE VASCONCELOS, M.C.B., BENNETT, R.N.,
ROSA, E.A.S. and FERREIRA-CARDOSO, J.V. 2009. Industrial processing effects on chestnut fruits (Castanea
sativa Mill.). 2. Crude protein, free amino acids and phenolic
phytochemicals. Int. J. Food Sci. Technol. 44, 2613–2619. MÍGUELEZ, J.D.L.M., BERNÁRDEZ, M.M. and QUEIJEIRO,
J.M.G. 2004. Composition of varieties of chestnuts from Galicia (Spain). Food Chem. 84, 401–404.
MOREIRA, R., CHENLO, F., CHAGURI, L. and VÁZQUEZ, G. 2005. Mathematical modelling of the drying kinetics of chestnut (Castanea sativa Mill.). Influence of the natural shells. Food Bioprod. Process. 83, 306–314.
MOREIRA, R., CHENLO, F., CHAGURI, L. and FERNANDES, C. 2008. Water absorption, texture, and color kinetics of air-dried chestnuts during rehydration. J. Food Eng. 86, 584–594.
MOREIRA, R., CHENLO, F., CHAGURI, L. and MAYOR, L. 2011. Analysis of chestnut cellular tissue during osmotic
dehydration, air drying and rehydration processes. Drying Technol. 29, 10–18.
MOREIRA, R., CHENLO, F., TORRES, M.D. and RAMA, B. 2013. Influence of the chestnuts drying temperature on the rheological properties of their doughs. Food Bioprod. Process.
91, 7–13.
PEREIRA-LORENZO, S., RAMOS-CABRER, A.M., DÍAZ-HERNÁNDEZ, M.B., CIORDIA-ARA, M. and RÍOS-MESA, D. 2006. Chemical composition of chestnut cultivars from Spain. Sci. Hortic. (Amsterdam) 107, 306–314. RAJASEKAR, D., AKOH, C.C., MARTINO, K.G. and
MACLEAN, D.D. 2012. Physico-chemical characteristics of juice extracted by blender and mechanical press from pomegranate cultivars grown in Georgia. Food Chem. 133, 1383–1393.
TOGRUL, I.T. and PEHLIVAN, D. 2003. Modelling of drying kinetics of single apricot. J. Food Eng. 58, 23–32.
YALDÝZ, O. and ERTEKÝN, C. 2001. Thin layer solar drying of some vegetables. Drying Technol. 19, 583–597.
ZHANG, M., CHEN, H. and ZHANG, Y. 2011. Physicochemical, thermal, and pasting properties of Chinese chestnut (Castanea
mollissima BI.) starches as affected by different drying
